University of Virginia Library

I. THE INTELLECTUAL ANCESTRY
OF BEHAVIORISM

1. Aristotle. Many things in the history of thought
have been traced back to Aristotle with varying de-
grees of appositeness, but there is almost nothing ap-
posite in tracing back behaviorism to him. To credit
him, for instance, with taking up a position on the
central methodological issue of the use of publicly
observable data, as distinct from introspective evi-
dence, in studying human beings would display a gross
lack of historical perspective; for the distinction be-
tween the private world of the individual's own con-
sciousness and the public world, which all could ob-
serve, was alien to the Greeks. Indeed there is a sense
in which the Greeks had no concept of consciousness
in that they did not link together phenomena such as
pain, dreams, remembering, action, and reasoning
as exemplifying different modes of individual con-
sciousness. The concept of consciousness was largely a
product of individualism, of the various movements
such as Stoicism, Epicureanism, and Christianity,
which supplied types of conceptual schemes that were
very different from those which were appropriate to
the shared life of the city-states. The coordinating
concept of individual consciousness was not made ex-
plicit until it found expression in the systems of Saint
Augustine and Descartes. The use of introspection as
a technique for investigating consciousness went along
with such systems of thought, and behaviorism can only
be understood as a reaction against such a technique.
It would, therefore, be absurd to search for hints of
the central doctrine of behaviorism in a thinker such as
Aristotle, whose way of thinking about human life
antedated the conceptual schemes which permitted
such questions to be raised.

What can be said about Aristotle is that, being a
marine biologist by training, he was the first to ap-
proach the study of human beings in an objective and
systematic way. He developed a classificatory system
which included plants, animals, and man as belonging
to the same genus of living things. He sent his research
workers all over the known world to provide him with
facts, not only about the different species of living
things but also about the different types of customs
and systems of government under which men lived.
This was all recorded and fed into the classificatory
system that he developed at the Lyceum.

When, however, we turn to Aristotle's Metaphysics
and De anima, and study the conceptual scheme which
he thought appropriate for describing and explaining
human behavior, we find not just that his doctrine of
form and matter was incompatible with the materi-
alism espoused by many behaviorists but that, in his
psychology, he was an explicit critic of the mechanists
of his day.

Aristotle held that a living thing is a “body with
a soul,” “soul” designating the self-originated tendency
of living things to persist towards an end. This tendency
can be exhibited at the nutritive and reproductive level
as in plants, at the level of sensation and movement
as in animals, and at the rational level as in man.
Aristotle accused mechanists such as Democritus and
Empedocles of the all-pervading mistake of concluding
from the fact that the soul is the cause of movement,
that it is itself moved. He maintained that the soul
moves the body “by means of purpose of some sort,
that is thought.” Thinking is not a sort of motion any
more than desire or sensation are. His predecessors had
misunderstood the sort of concept that “soul” was.
Insofar as it is a capacity, how could it be moved?
A capacity is not the sort of thing that can be moved.
Insofar as it is an exercise of a capacity, such as think-
ing, it is manifest in a process that cannot adequately
be described as a change in motion. Aristotle deployed
many ingenious arguments to substantiate this criticism
of mechanical theories, many of which are similar to
those which can be found in the work of modern
philosophers such as Ryle (Peters [1962], pp. 102-04;
Ryle [1949]).

There are thus almost no grounds for linking Aris-

totle with behaviorism either in respect of its central
doctrine or in respect of its more peripheral ones. If
Aristotle is to be linked with any school in twentieth-
century psychology the obvious one would be that of
the “hormic” (purposive) psychology championed by
William McDougall. For here too we find behavior
studied objectively, an exaltation of purpose as the
most important explanatory concept, and a vehement
attack on the mechanists of his day, namely J. B.
Watson and the reflexologists. Indeed McDougall's
indebtedness to Aristotle is explicitly acknowledged at
many points.

2. Hobbes. There would be more plausibility in
attempting to trace behaviorism back to Hobbes than
to Aristotle. To start with, Hobbes was one of the great
thinkers of individualism and wrote at a time when
the private world of the individual was both recognized
and valued—and threatened by tendencies towards
absolutism. Hobbes himself regarded man's capacity
to form “phantasms” or images as one of his most
miraculous powers. “Of all the phenomena or appear-
ances which are near to us, the most admirable,” he
says “is apparition itself, τὸ αάινεσθαι ; namely, that
some natural bodies have in themselves the pattern
almost of all things, and others of none at all” (Hobbes
[1839a], p. 389). It was man's mysterious power to
register within himself what was going on around him
and to store up his impressions for use on further
occasions that awakened Hobbes's passionate curiosity.
How could this mysterious power be explained? This
was the problem that lay at the heart of Hobbes's
psychology and theory of nature.

Thus Hobbes's starting point in psychology reveals
both the conceptual possibility of behaviorism for him
and also the absurdity of thinking that, in the most
important respect, Hobbes was in fact a behaviorist;
for no behaviorist could regard the problem of imagery
as the most important phenomenon for a psychologist
to explain. It is also difficult to see how much could
be done about explaining it without constant resource
to introspection.

On the central question of the appropriate data for
a science of human behavior Hobbes was, as a matter
of fact, absolutely explicit. In his Introduction to
Leviathan he wrote:

(Hobbes, 1839b, Introduction).

Hobbes not only extolled introspection as the appro-
priate method for investigating mankind; he also
pointed to the unreliability of inferences made on the
basis of the observation of others. Since Hobbes ac-
cepted the use of introspective evidence, why has his
linkage to behaviorism seemed so plausible to so many?
There are, first of all, some underlying assumptions
which are common to the views of Hobbes and modern
behaviorists, and these are so deeply embedded in
modern thought that we tend to take them for granted.
The first is the assumption that there is some reliable
method for advancing knowledge. Hobbes was one of
the many “new men” of the post-Renaissance period
who believed that knowledge of nature was available
to anyone who was prepared to master the appropriate
method. He thought that Copernicus and Galileo had
revealed the method for investigating the natural
world, that Harvey had applied this to the study of
the body, and that he, Hobbes, was showing how this
method, the resoluto-compositive method of Galileo,
could be applied to psychology and politics.

Hobbes's early contact with Francis Bacon, for
whom he had worked for a period as a kind of literary
secretary, had also convinced him that knowledge
meant power. Hobbes's psychology and politics were
constructed with a very practical end in view—the
preservation of peace, and he thought that there was
no hope for England, in the throes of civil war, unless
those who had some influence on the course of events,
could be persuaded to accept the logic of his demon-
strations concerning man and civil society. This practi-
cal concern underlying his theorizing, which was later
to be applauded by Marxists, was another underlying
link between Hobbes and the behaviorists.

A much more explicit link between Hobbes and the
behaviorists was his materialism, and his attempt to
extrapolate the concepts and laws of Galileo's me-
chanics to the human sphere. “For seeing life is but
motion of limbs... what is the heart but a spring;
and the nerves but so many strings; and the joints but
so many wheels, giving motion to the whole body, such
as was intended by the artificer” (Hobbes, 1839b,
Introduction). Desires and aversions are motions to-
wards and away from objects. Thinking is but motion
in some internal substance in the head and feeling is
movement about the heart. Imagery, which he found
so wonderful, was to be viewed as a kind of meeting
place of motions. The phenomena of perception and
imagination could be deduced from the law of inertia.
In order to make such deductions Hobbes postulated
infinitely small motions, which he called “endeavours,”
in the medium between the object of sense and the
brain, and he had recourse to them also to explain how
movements coming from outside bodies are passed on
through the body so that they eventually lead to the

gross movements observable in desire and aversion.

Within behaviorism it is customary, following
Tolman, to distinguish between molecular and molar
theories of behavior. A molecular theory, such as that
of Clark Hull, is one which starts from postulates at
the physiological level and attempts to deduce the
movements involved at the molar level, e.g., the gross
movements of the body, from them. Hobbes antici-
pated such molecular theories to an astonishing extent
(Peters and Tajfel, 1957). But such anticipation had
nothing to do with behaviorism in a strict sense. It was
rather the consequence of applying the hypothetico-
deductive procedure of Galileo, together with his me-
chanical concepts and laws, to the realm of human
behavior. Hull, combined this Galilean approach to
psychology with the restriction of data to what could
be publicly observed, which was the central feature
of behaviorism (Hull, 1943). Hobbes, therefore, can
properly be regarded as the father of mechanistic
theories in psychology rather than of behaviorism; for
not all behaviorists were mechanists, and Hobbes him-
self relied on introspection in the psychological sphere.

3. Descartes. It would be more absurd to regard
Descartes as a behaviorist than Hobbes; for he was
notorious for the dualism of mind and matter which
he postulated. But, as a matter of fact, both his dualism
and his assumptions about scientific method did much
to create the climate of opinion which made behav-
iorism possible, if not almost inevitable.

Descartes held that there are two sorts of substances
in the world, mental and physical. If the behavior of
these substances was to be scientifically studied, as-
sumptions about them had to be made explicit and
exhaustively analyzed until clear and distinct ideas
were arrived at, which were simple in the sense that
no further analysis of them was possible. In the case
of ideas about material objects, for instance, the scien-
tist eventually arrived at the simple ideas of extension,
figure, and motion. If certain of these simple ideas were
combined, relationships could be grasped between
them which served as postulates for a deductive system,
as in geometry. Thus the understanding of bodies and
of minds respectively rested upon clear and distinct
ideas which had no features in common. Descartes'
problem about the relationship between mind and body
derived from the fact that, though in our confused
everyday experience we are aware of interaction, as
when our limbs move because of our intentions, no
clear and distinct idea can be formed of the manner
of this union. Such perspicuous ideas are only forth-
coming in the spheres of the mental and the physical
when they are proceeding independently of each
other—as in logical reasoning on the one hand or in
reflex movements on the other.

Descartes' dualism and his assumptions about scien
tific method thus gave rise to two traditions of enquiry
which came to be pursued more or less independently
of each other. On the one hand the human body, which
was regarded as functioning mechanically right up to
the level of instinctive behavior and simple habits, be-
comes a fit subject for objective study. Harvey had
made a splendid advance in this field with his mechan-
ical theory of the circulation of the blood. On the other
hand, the mind, by which Descartes meant mainly the
higher thought processes and the will, could only be
studied introspectively. The consequence of Descartes'
dualism was, therefore, the school of mechanistic biol-
ogy and reflexology on the one hand and the intro-
spective school of psychology on the other, which
reached its culmination about 250 years later in the
laborious experimental work of Wundt and Titchener.

It was against the assumptions of the introspective
school that Watson revolted—their assumptions about
both introspective method and the “stuff” of conscious-
ness which he claimed they were trying to study by
this method. And when he revolted he fell back on
the other tradition stemming from Descartes: mecha-
nistic biology and reflexology. All he did was to at-
tempt to extend its domain to the level of thought and
action which had previously been regarded as “mental”
and hence to be studied by introspective methods. And
when Watson theorized about behavior he was un-
wittingly Cartesian in his approach. He thought that
the complex phenomena of behavior could be ex-
plained by analysis into clear and distinct units of
behavior—simple reflexes.

4. Reflexology. Descartes' dualism involved the as-
sumption that the behavior of the body, below the level
of willed action, could be explained mechanically. He
had, however, a crude idea of how the body works.
He thought of it as a statue or machine made of earth
and was much impressed by the feats which mechanical
manikins could be made to perform in the gardens
of the aristocracy by arranging water-pipes within
them. They could be made to move their limbs and
even to produce sounds like words. He thus pictured
the nervous system as a piece of intricate plumbing.
The nerves were thought of as tubes along which
“animal spirits,” which occupied an indeterminate
status between the mental and the physical, flowed
continuously. Changes in the motion of these spirits
cause them to open certain pores in the brain. When
this happens the motion of the animal spirits is changed
and they are “reflected” into the muscles which move
the body. For he thought that many movements of the
body are not brought about by conscious intention, but
by an undulatio reflexa, or a movement of rebound
in the animal spirits at the meeting of the sensory and
motor channels at the pineal gland, where the mind
could also influence the body by means of images. The

automatic reactions of the body, which were not under
voluntary control, were thus called reflexes.

Little was done to refine this conception until 1811
when Charles Bell published a paper entitled “An Idea
of a New Anatomy of the Brain,” which he communi-
cated to the Royal Society in 1821, and in which he
claimed that the nerves, which are connected with the
spinal centers by anterior roots, are employed in con-
veying motor impulses from the brain outward, and
that the sensory nerves are connected with the poste-
rior roots of the spinal cord. This was confirmed by
Magendie in 1822. In 1833 Marshall Hall demonstrated
clearly the existence of reflex action which proceeded
independently of conscious volition, and in the latter
part of the nineteenth century the antics of animals
deprived of their higher brain structure were a com-
monplace. In 1851 Claude Bernard pioneered physio-
logical work on the influence of specific nerves on the
blood vessels and the consequent changes throughout
the sympathetic system, thus helping to understand the
connection between the brain and the viscera and other
changes involved in emotional and motivational states.
Evolutionary theory, especially that of Herbert
Spencer, led Hughlings Jackson to postulate different
levels of evolution in the nervous system from the less
to the more organized, from the automatic to the
voluntary.

From the point of view of the history of behaviorism
the crucial step forward was taken by Pavlov, whose
particular interest was in the digestive system. In 1897
he published a book on The Work of the Digestive
Glands, in which he noted that there are certain ir-
regularities and interruptions in the work of these
glands, which he attributed to psychic causes, e.g., that
sometimes the glands would start to work before food
was given to a dog, when the dog saw the man who
usually fed it. In 1902 he embarked on a long series
of experiments to study such phenomena. He concen-
trated on salivation, rather than on gastric secretion,
because it was more accessible to experimental analy-
sis. A dog was strapped in a test frame, with elaborate
experimental controls, and a bell (conditioned stimulus)
was repeatedly sounded before food (unconditioned
stimulus) was placed in the mouth to produce salivation
(unconditioned response), until eventually the sound
of the bell brought about salivation (conditioned re-
sponse) before the presentation of the food. Pavlov also
found that the conditioned stimulus becomes general-
ized, in that the dog comes to respond to a wide range
of stimuli. He found, too, that dogs could be taught
to discriminate between stimuli by rewarding responses
to one stimulus, such as a circle, but not to another,
such as an ellipse. If the difference between the stimuli
was gradually reduced a point would be reached where
the dog's behavior evinced all the symptoms of acute
neurosis. The concept of “reinforcement” was invented
to refer to this process in which the conditioned stimu-
lus is presented in close juxtaposition to the uncondi-
tioned stimulus. Many have commented on the simi-
larity between this concept and that of Thorndike's
“reward,” central to his law of effect. But the two
concepts emerged from very different theoretical
backgrounds and their differences are as important as
their similarities (Wolman [1960], pp. 53-55).

Pavlov was unrepentantly a physiologist and he
linked his experimental findings with a theory about
irradiation and processes of excitation and inhibition
in the brain. He expressed contempt for psychology
and refused to take sides in psychological controversies.
Nevertheless his influence has been nugatory in physi-
ological theory but vast in psychology, because the
behaviorists later seized upon his findings. His contem-
porary, Bekhterev, on the other hand, who also popu-
larized the conditioned reflex, was more catholic in
his interests. In 1907 he published his Objective Psy-
chology in which he proclaimed that the future of
psychology depended upon objective, external obser-
vation. He envisaged excluding introspective data and
mentalistic concepts and basing psychology on physical
and physiological findings. In this respect Bekhterev
harped back to La Mettrie and the materialistic tradi-
tion in psychology. In his conditioning experiments he
did not confine himself to reactions such as salivation
but met with some success in conditioning motor re-
sponses as well. He also took an interest in speech,
as also did Pavlov towards the end of his life.

Watson embarked upon his behavioristic program
in ignorance of the physiological studies of Pavlov and
Bekhterev, but he gradually incorporated them into
his theory when he became familiar with them through
translation. Thus Watson's reflexes functioned in his
theory as the direct descendants of Descartes' “simple
natures” in the bodily sphere. The links between them,
however, namely the principles of association, and the
assumptions about how generalizations could be ar-
rived at about such links, came from another source—
the empiricist tradition. A brief exposition of the lead-
ing ideas of this tradition will complete the account
of behaviorism's intellectual ancestry.

5. The Empiricist Tradition. The other aspect of
Descartes' thought, the interest in the contents of con-
sciousness, was developed by both rationalists and
empiricists alike. The empiricists, like the rationalists,
were really preoccupied with problems of knowledge.
As John Locke put it, they were concerned with the
“original, extent, and certainty” of human knowledge.
They held, however, that knowledge was based on
experience, not in the unexceptionable sense that how-

ever we come to obtain beliefs about the world, their
truth or falsity must be tested by comparing them with
what can be observed, but in the much more dubious
sense that our ideas about the world originate in our
own individual sense-experience. Therefore, questions
about the extent and certainty of knowledge tended
to resolve themselves into speculations about how ideas
originated; for followers of what was called “the way
of Ideas” held that genuine ideas must be tracked back
to impressions of sense. The result was that philo-
sophical questions about the meaning of terms and
about the grounds of knowledge were systematically
confused with questions in genetic psychology about
their origin. It was not until the nineteenth century,
when F. H. Bradley proclaimed that “In England we
have lived too long in the psychological attitude”
(Bradley [1922], 1, x) that this confusion, which per-
sisted from Locke to James Mill and Alexander Bain,
began to be systematically exposed.

The net result of this confusion was that the work
of the empiricists contained both a philosophical the-
ory about the grounds and acquisition of knowledge
and a psychological theory about the working of the
mind. The philosophical theory came straight from
Francis Bacon. To obtain knowledge, it was held, a
start had to be made with simple uninterpreted data,
or impressions of sense. There must be no premature
hypotheses or “anticipations of Nature.” Generaliza-
tions had to be made which reflected regularities in
the data. Bacon elaborated tables of co-presence, co-
absence, and co-variation of instances to ensure that
these generalizations were well-founded. These were
later elaborated by Mill in his celebrated methods of
experimental enquiry. It was of cardinal importance
in this process of “induction” that generalizations
should not go beyond the data and that no recourse
should be made to unobservables. Laws expressed
correlations between what could be observed.

The psychological theory which developed pari-
passu with this philosophical theory about the grounds
and acquisition of knowledge had two main features.
Firstly it maintained that the experience, thought, and
consequent action of the individual is caused from
without. The environment causes simple ideas (Locke)
or impressions (Hume) to arise in the individual. The
individual's body was also regarded as part of the
external world which gives rise to impressions of
reflexion—e.g., of pleasure and pain—which enter the
mind through different types of receptor. (This was
later on called “the inner environment” by behavior-
ists.) Secondly it was held that the ideas arising from
these two environmental sources become linked to-
gether by principles of association such as contiguity
and resemblance. Action is initiated by an idea that
has become linked with pleasure or pain. Thus food,
for instance, gives rise to an idea in the mind, which
has become linked to the idea of eating, which in its
turn has been linked with the idea of pleasure. This
brings about the action of eating the food. This account
of the initiation of action can be found in Hobbes,
though he did not give much prominence to the associ-
ation of ideas in his account of thinking. He stressed
the importance of plans deriving from desire.

The history of empiricism is largely the history of
the elaboration and sophistication of these basic ideas.
In the philosophical and methodological sphere there
were three main derivative doctrines. Firstly the notion
of “data” was gradually sophisticated into modern
theories of sense-data, which can be found in the work
of philosophers such as G.  E . Moore, Bertrand Russell,
and A. J. Ayer. In the scientific sphere Kant did much
to popularize the idea that the domain of science was
coextensive with the domain of the measurable. It
became important, therefore, for scientists to obtain
data which were as precise and pure as possible by
devising various forms of measurement. In psychology
the nineteenth-century concentration on psycho-
physics, stemming largely from the work of Weber and
Fechner, bore witness to this search for measurable
data—the notorious just noticeable differences.

Secondly, a theory of meaning developed which has
come to be known as “logical empiricism.” This main-
tained that only those terms are strictly meaningful
that can be cashed by reference to what can be ob-
served. The language of morals and of poetry is, strictly
speaking, meaningless (or has merely “emotive mean-
ing”) because it cannot be tied down in this way to
observables. Scientific terms have either to be “opera-
tionally defined” or related indirectly to observation
by a process of “logical construction.” In the early
twentieth century P. W. Bridgman's book The Logic
of Modern Physics (1927) popularized this view of
scientific terms. Its leading ideas were applied to psy-
chology by C. C. Pratt in The Logic of Modern Psy-
chology (1939) and exerted a considerable influence on
B. F. Skinner, a leader of modern behaviorism.

The third development was the attempt to formulate
precisely the methods for arriving at well-founded
generalizations and to get clearer about the theory of
probability which was presupposed. The works of J. S.
Mill, W.  E . Johnson, and J. M. Keynes were classics
in this tradition.

The psychological theory which was favored by most
of the empiricists was given an ambitious start by
David Hume who pictured himself as the Newton of
the sciences of man. Simple impressions were regarded
as mental atoms and the principles of association were
postulated as performing, in the mental sphere, the

same function of uniting them together as was per-
formed by the principle of gravitational attraction in
the physical sphere. David Hartley developed an even
more ambitious version of this type of theory; for he
held that the psychological principles of association
paralleled the mode of operation of physiological
disturbances in the substance of the nerves, spinal
marrow, and brain, which he called “vibrations.” It
was left to James Mill, however, to free associationism
from Newtonian pretensions and physiological specu-
lation and to attempt to formulate soberly and prosa-
ically the basic principles in terms of which ideas were
thought to be connected. Most of the subsequent work
in the nineteenth century of the British associationist
school consisted in criticism, refinements, and simpli-
fications of Mill's edifice.

In France, largely through the infectious cynicism
of Voltaire, British empiricism came to exert an in-
fluence that was more mundane, and less theoretical.
It encouraged thinkers to observe more carefully and
more objectively how men in fact behaved. Diderot's
Lettre sur les aveugles and Lettre sur les sourds et muets
were classics of their kind—concrete case studies of
individual lives. Similarly Condillac approached
Locke's problems in a more concrete, if more imagina-
tive way, by creating the fiction of a statue endowed
only with the faculty of smell. And Cabanis, a vehe-
ment critic of Condillac, began his psychophysiological
writings in 1795 with an attempt to answer the con-
crete, if depressing question, of whether the victims
of the guillotine suffer any pains after decapitation.
His theory, which attacked Condillac's starting point
of imagining a being capable of sensation in isolation
from the structure of the organism as a whole, was
diametrically opposed to the atomism of the associa-
tionist tradition. But it was a theory based on actual
observations of men from childhood to maturity. Simi-
larly, La Mettrie, who elaborated Hobbes's thesis that
man is a superior type of machine, developed his the-
ory not as an imaginative extrapolation from Galilean
mechanics but partly as a result of his medical studies
under Hermann Boerhave, and partly from direct ob-
servation of his own experiences during a fever. And
in the nineteenth century Taine, who represented very
much the antimetaphysical, positivistic school in
France, scrambled together, in his De l'intelligence
(1870; trans. as On the Intelligence, 1871), reports from
asylums, physiological facts, and references to Mill's
Logic!

It would be tempting to suppose that this interest
in the concrete behavior of men, and the attempt to
study it objectively, which was so characteristic of
French empiricism, was one of the formative influences
in the development of behaviorism. There is, however,
little plausibility in this suggestion. For the rise of
behaviorism is to be explained partly as a method-
ological reaction to introspective psychology and
partly as a consequence of the success which was being
attained in the study of animals. Almost the last thing
which the behaviorists actually came to study was the
concrete behavior of men. Let us now pass, therefore,
to the immediate origins of behaviorism.